Work from a number of laboratories, including those involved in this program project, has demonstrated that Adeno-associated virus (AAV) holds significant promise for the correction of human diseases, This work has shown that AAV can be used to transfer genes efficiently into primary cells in vivo, and that in most cases, expression of the transgene appears to be long lived. Additionally, the work done in this program project has led to the development of new and potentially scalable methods for growing rAAV, as well as methods for purifying rAAV that produce high titer recombinant virus that is free of wild type virus or other contaminants. Much of this work has focused on AAV2, a serotype that shows a broad host range and a broad tropism with respect to the types of differentiated cells that it can transduce. Recent work with other AAV serotypes suggests that they too have a broad although somewhat different tropism. Although a broad host range is useful, it is clearly time to begin exploring ways of developing AAV vectors that have a more restricted or specific tropism, or vectors that have special properties. In particular, it would be extremely helpful if methods could be found to target AAV vectors to specific tissues. However, vector targeting is still in its infancy and is particularly hampered in the case of AAV by the relative lack of information about capsid assembly, particle entry and intracellular trafficking. To facilitate our ongoing studies of targeting, it will be necessary to understand the basic biology of AAV capsid structure. To address these problems, this proposal focuses on structure function studies of the AAV2 capsid genes and on the determination of the crystal structure of the two most dissimilar AAV serotypes when compared to AAV2, namely AAV4 and 5. The specific aims are: 1) We will use site specific mutagenesis to identify the amino acids that are potentially involved in maintaining capsid integrity at the 2 fold and the 5 fold axes of symmetry. 2) We will identify regions of the capsid proteins required for nuclear localization. 3) We will determine the atomic structure of AAV4 and AAV5 using X-ray crystallography and map the sialic acid binding regions using cryoelectron microscopy. It is anticipated that these studies will produce valuable information that will impact on the use of AAV vectors for virtually all gene therapy studies.